The fact that fear of victimization is relatively high among the elderly while rates of victimization are relatively low suggests that fear of victimization, to some degree, varies independently of the actual risk of victimization. The objective of this study is to determine antecedents of fear of victimization other than victimization itself or the risk thereof. The basic hypothesis is that fear of victimization varies inversely with social integration. Data have been obtained from a sample of over 3700 residents of Washington State aged 55 and over. Respondents were located by a telephone survey employing random digit dialing methods and screened for eligibility. Permission to send a mailed questionnaire was requested from those aged 55 or over; 81.5% of the eligible persons agreed to receive the questionnaire. The response rate to the questionnaire is approximately 74% at this point, although this stage of the data-collection process is not quite completed. The next year of the project will be devoted to data analysis and hypothesis testing. Theoretical models employ demographic, economic, and health-related variables as antecedents of social integration factors, which are in turn antecedent to fear of victimization.